Many patients with epilepsy are resistant to standard anticonvulsant drug (ACD) treatments. This proposal seeks to elucidate the origin of the genetic factors that affect individual response to ACDs by mapping the location of genes that influence these responses in mice. There are 2 phases to this proposal. Phase 1 involves identifying mouse strains best suited for dissecting the genetic influences which control response to specific ACDs. Phase 2 involves mapping these genetic influences to defined regions of the genome. In phase 1, strain-specific maximal electroshock seizure threshold (MEST) will be characterized. Strains with similar mean MESTs will be considered equivalently seizure-sensitive and such pairs of strains will be used for subesequent anticonvulsant drug (ACD) testing. ACD testing will involve dose-reponse studies with a panel of clinically relevant ACDs: phenytoin, carbamazepine, vaiproic acid and gamma-vinyl GABA. The quantitatve endpoint will be the absolute MEST determined in the presence of drug. Strains will be selected for quantitative trait loci (QTL) analyses based on their strain-specific response such that pairs of strains exhibiting the largest differential effects on MEST for a given drug (the strains showing the largest and smallest anticonvulsant effects) will be used for QTL studies. Brain levels of ACDs will be determined in parental strains in order to address one possible major co-phenotype in correlation with the anticonvulsant MEST response. In phase 2, QTL mapping studies will be conducted using mouse strains suggested by phase 1 phenotype studies. Mapping will utilize segregating F2 (intercross) populations for each strain pair. Quantitative phenotype for mapping will be MEST in individual F2 mice pretreated with a specific ACD. Brain ACD levels will be determined in F2 animals and used as a second quantitative phenotype for mapping. In order to distinguish ACD response QTLs from seizure sensitivity QTLs which may segregate in the cross, a parallel QTL study will be conducted for each ACD using an independent F2 population tested for MEST following saline rather than ACD pretreatement. QTL genotype and mapping experiments will combine a 15-20 cM genome scan with comprehensive statistical analyses including both parametric and non-parametric single and multilocus models. Results will lead to the direct localization of genes that influence anticonvulsant responses in mice with future directions involving the identification of these genes. The described studies build directly from the foundation of work in the investigator's lab on mapping mouse loci involved in differential sensitivity to chemically- and electrically-induced seizures and ultimately will lead to a focused strategy for investigating genetic influences on response to anticonvulsant drugs in humans with epilepsy. The association of human anticonvulsant response with specific genomic variants will lead to more rational decisions regarding the choice of drug for individual patients and will lead to greater success in treating seizures disorders in general.